Re: [akka-user] Re: akka-http: combine path together
I was confused. I thought you were looking for something like:
path(v2.0 / tokens) {
post {
complete(it works)
}
} ~
path(v2.0 / somethingelse) {
post {
complete(it works)
}
}
Otherwise with
path(v2.0) {
path(tokens) {
You're matching against path(v2.0) and then further trying to also match
against path(tokens). Since a path can't start with both v2.0 and
tokens at the same time, this always fails. There are other path
directives you can use for this type of processing and I think you've
already discovered some of them.
akka-http is based off of spray concepts. Spray's documentation is
currently much better than akka-http. You should probably start
here http://spray.io/documentation/1.2.2/spray-routing/ to learn more about
what routing directives are available and how to compose them together.
All directives may not exist yet in akka-http, but the sections on the
Path DSL should be valid.
--
Read the docs: http://akka.io/docs/
Check the FAQ:
http://doc.akka.io/docs/akka/current/additional/faq.html
Search the archives: https://groups.google.com/group/akka-user
---
You received this message because you are subscribed to the Google Groups Akka
User List group.
To unsubscribe from this group and stop receiving emails from it, send an email
to akka-user+unsubscr...@googlegroups.com.
To post to this group, send email to akka-user@googlegroups.com.
Visit this group at http://groups.google.com/group/akka-user.
For more options, visit https://groups.google.com/d/optout.
[akka-user] [akka-http] an empty response body is send with the text/plain content type
Hi, by using complete(HttpResponse(status = NoContent)) or complete(NoContent) the HTTP response contains the header: Content-Type: text/plain; charset=UTF-8 If I understand the code correctly, HttpEntity.Empty should not send this content type: val Empty: Strict = Strict(ContentTypes.NoContentType, data = ByteString.empty) val NoContentType = ContentType(MediaTypes.NoMediaType) Can somebody explains me where does this header come from? Cheers, Yann -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
[akka-user] Re: getting scala-zeromq (and/or akka-zeromq) to work with c++ program
Found the solution, just in case anyone gets in the same spot as I was
earlier, the problem is with the incompatibility between zmq 2.2 and zmq
4.0.4, although zmq website advertises zmq latest version stating it is
backwards compatible it is NOT. when I stepped down my zmq version to 3.2.5
my scala and c++ zmq programs started talking.
On Monday, March 9, 2015 at 3:50:54 PM UTC-7, ankit master wrote:
Hello,
I am trying to get c++(publisher) to my scala(subscriber ) with little
luck, I tried using both akka-zeromq and scala-zeromq to get this working
with little luck. however, my program *does work* with *scala being the
publisher and c++ being the subscriber (but not other way around)*, can
anyone please help me figure out what I am doing wrong here !!
The following code will not work, I dont get any messages over on scala
side, will be very grateful of any help here.
/ Scala Code , zmq version 2.2
import akka.actor._
import org.zeromq.ZMQ
import org.zeromq.ZMQ._
object ZMQWithoutAkkaExtensionApp extends App {
val numberOfMessages = 10
val system = ActorSystem(zmq)
val subscriber = system.actorOf(Props[Sub])
subscriber ! start
}
class Sub extends Actor {
var contextZMQ: ZMQ.Context = null
var subscriberSocket: ZMQ.Socket = null
override def preStart = {
contextZMQ = ZMQ.context(1)
subscriberSocket = contextZMQ.socket(ZMQ.SUB)
subscriberSocket.connect(tcp://localhost:1234)
subscriberSocket.subscribe(.getBytes())
Thread.sleep(1000)
}
def receive = {
case msg = handleMessages
}
private def handleMessages = {
while (true) {
println(-)
val request = subscriberSocket.recv(0)
println(srequest : $request)
val data = new String(request)
println(sdata: $data )
}
}
}
C++ Code , zmq version 4.0.4
#include zmq.hpp
#include string
#include iostream
#include windows.h
#include thread
using namespace std;
void Pub()
{
cout Running Pub endl;
zmq::context_t context(1);
zmq::socket_t publisher(context, ZMQ_PUB);
publisher.bind(tcp://*:1234);
Sleep(4000);
cout awake again endl;
int cc = 0;
while (1) {
// Write two messages, each with an envelope and content
string msg = hello ;
zmq::message_t req(5);
memcpy((void *)req.data(), World, 5);
publisher.send(req);
cout cc++ endl;
Sleep(1000);
}
}
int main()
{
Pub();
return 0;
}
Thank you
Sincerely,
AVM
--
Read the docs: http://akka.io/docs/
Check the FAQ:
http://doc.akka.io/docs/akka/current/additional/faq.html
Search the archives: https://groups.google.com/group/akka-user
---
You received this message because you are subscribed to the Google Groups Akka
User List group.
To unsubscribe from this group and stop receiving emails from it, send an email
to akka-user+unsubscr...@googlegroups.com.
To post to this group, send email to akka-user@googlegroups.com.
Visit this group at http://groups.google.com/group/akka-user.
For more options, visit https://groups.google.com/d/optout.
Re: [akka-user] Re: Problems using a Remote Actor
Hi Jim, thanks for the help ... Have you followed all of the instructions for interacting with remote actors here: http://doc.akka.io/docs/akka/2.3.9/scala/remoting.html ? I follow this because I'm using the Java version of the API: http://doc.akka.io/docs/akka/2.2.5/java/remoting.html and them the updated (more/better info) http://doc.akka.io/docs/akka/2.3.9/java/remoting.html Note that I have the same behavior from 3 different pc (Windows and Linux) and with different patch levels on JDK 7 (but all recent), so it's something related to code ... Just for info, my project is made with Gradle 2.3, so to try it its' enough to checkout with git, and run it with Gradle (if already installed, or its wrapper will download it), and all dependencies are downloaded at first run, I made it to make it simple for anyone to run/try it. From the error it appears to me that: 1. You are attempting to create a remote actor from a peer yes, but I didn't read for problems with it in documentation ... 2. This peer isn't able to properly interact with [akka.tcp:// RemoteActorSystem-Client@127.0.0.1:55003] yes but from the error/warning message it's not clear what is the real problem ... Are you using a Remote or Clustered ActorRefProviders on both nodes? no, I'm using normal akka.remote.RemoteActorRefProvider Is your remote transport properly configured to run on those ports? Is it starting properly? Is your networking configured to allow access to those ports (software firewalls on a node have burned me in the past). yes, I see that the related port is open, with an OS diagnostic tool Does the RemoteActorSystem-Client AS exist on that port, or is some other AS running there? nothing special is running There's a lot to look at, but it ought to work once everything is wired properly. yes, but all seems ok, so really I think it's a problem in my code Maybe look for a working example (from Activator) and then adapt the working thing to your needs? maybe this is a good option, but my sample here is even to try to use Akka from a (little) different point of view (and tools like Gradle) ... maybe even Akka documentation could be improved, at least to better explain some pitfalls on exposing actors with remoting ... just as a suggestion for Akka team. One of goals for my test project is even to share/improve some knowledge on using Akka from other tools (mainly the Groovy/Gradle/Grails stack) with others :-) . Thanks a lot for the help. Sandro -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
[akka-user] Re: Akka http - complete with headers
oh! That was it! thanks ;)
I'll remember to check if they add it to the documentation once it's
released.
Cheers
G
On Thursday, 12 March 2015 10:00:46 UTC, Mark Hatton wrote:
Completing a route with a Tuple3[StatusCode, Seq[HttpHeader], T] doesn't
seem to work for me.
Ensure your headers collection is an instance of immutable.Seq. E.g.:
import collection.immutable._
complete( (OK, Seq(`Cache-Control`(`no-cache`)), responseEntity) )
Works for me.
Mark
On Thursday, March 12, 2015 at 12:18:28 AM UTC, Giovanni Alberto
Caporaletti wrote:
This is what I came up with (I have another question below):
def onSuccessHead(magnet: OnStreamSuccessHeadMagnet): Directive[magnet.Out]
= {
magnet.directive
}
trait OnStreamSuccessHeadMagnet {
type Out
def directive: Directive[Out]
}
object OnStreamSuccessHeadMagnet {
import akka.http.server.directives.FutureDirectives._
implicit def apply[T](s: Source[T, _])(implicit mat:
ActorFlowMaterializer, tupler: Tupler[T]) = {
new OnStreamSuccessHeadMagnet {
type Out = tupler.Out
val directive =
onSuccess(s.runWith(Sink.head(SourceMarshallerHeadSink)): Future[T])
}
}
}
and the route:
onSuccessHead(userService.createUser(req)) { newUser =
respondWithHeader(Location(s/$v1/$user/${newUser.id})) {
complete(Created - newUser)
}
}
By the way, is there a way of extracting constant paths? e.g. my path
is path(v1 / user) but I don't want to have separate constants for v1
and user and something like path(some / path) { (some, path) = ...}
doesn't work. In other words, how do you do reverse routing in spray/akka
http?
Thanks
G
On Thursday, 12 March 2015 00:10:24 UTC, Giovanni Alberto Caporaletti
wrote:
Hi,
Completing a route with a Tuple3[StatusCode, Seq[HttpHeader], T]
doesn't seem to work for me. My json4s marshaller serializes the whole
tuple as an object. Tuple2[StatusCode, T] works instead.
Are there any plans to implement it?
What's the best way to add headers to a response anyway? In my case the
header depends on a future (or, well, on a materialization of a stream).
Thank you
Cheers
G
--
Read the docs: http://akka.io/docs/
Check the FAQ:
http://doc.akka.io/docs/akka/current/additional/faq.html
Search the archives: https://groups.google.com/group/akka-user
---
You received this message because you are subscribed to the Google Groups Akka
User List group.
To unsubscribe from this group and stop receiving emails from it, send an email
to akka-user+unsubscr...@googlegroups.com.
To post to this group, send email to akka-user@googlegroups.com.
Visit this group at http://groups.google.com/group/akka-user.
For more options, visit https://groups.google.com/d/optout.
[akka-user] Re: Akka http - complete with headers
Completing a route with a Tuple3[StatusCode, Seq[HttpHeader], T] doesn't
seem to work for me.
Ensure your headers collection is an instance of immutable.Seq. E.g.:
import collection.immutable._
complete( (OK, Seq(`Cache-Control`(`no-cache`)), responseEntity) )
Works for me.
Mark
On Thursday, March 12, 2015 at 12:18:28 AM UTC, Giovanni Alberto
Caporaletti wrote:
This is what I came up with (I have another question below):
def onSuccessHead(magnet: OnStreamSuccessHeadMagnet): Directive[magnet.Out] =
{
magnet.directive
}
trait OnStreamSuccessHeadMagnet {
type Out
def directive: Directive[Out]
}
object OnStreamSuccessHeadMagnet {
import akka.http.server.directives.FutureDirectives._
implicit def apply[T](s: Source[T, _])(implicit mat: ActorFlowMaterializer,
tupler: Tupler[T]) = {
new OnStreamSuccessHeadMagnet {
type Out = tupler.Out
val directive =
onSuccess(s.runWith(Sink.head(SourceMarshallerHeadSink)): Future[T])
}
}
}
and the route:
onSuccessHead(userService.createUser(req)) { newUser =
respondWithHeader(Location(s/$v1/$user/${newUser.id})) {
complete(Created - newUser)
}
}
By the way, is there a way of extracting constant paths? e.g. my path is
path(v1 / user) but I don't want to have separate constants for v1 and
user and something like path(some / path) { (some, path) = ...}
doesn't work. In other words, how do you do reverse routing in spray/akka
http?
Thanks
G
On Thursday, 12 March 2015 00:10:24 UTC, Giovanni Alberto Caporaletti
wrote:
Hi,
Completing a route with a Tuple3[StatusCode, Seq[HttpHeader], T] doesn't
seem to work for me. My json4s marshaller serializes the whole tuple as an
object. Tuple2[StatusCode, T] works instead.
Are there any plans to implement it?
What's the best way to add headers to a response anyway? In my case the
header depends on a future (or, well, on a materialization of a stream).
Thank you
Cheers
G
--
Read the docs: http://akka.io/docs/
Check the FAQ:
http://doc.akka.io/docs/akka/current/additional/faq.html
Search the archives: https://groups.google.com/group/akka-user
---
You received this message because you are subscribed to the Google Groups Akka
User List group.
To unsubscribe from this group and stop receiving emails from it, send an email
to akka-user+unsubscr...@googlegroups.com.
To post to this group, send email to akka-user@googlegroups.com.
Visit this group at http://groups.google.com/group/akka-user.
For more options, visit https://groups.google.com/d/optout.
Re: [akka-user] Akka Streams 1.0 release date?
Hi Ronald and Patrik, Are the release cycles of Akka Streams and Akka HTTP roughly the same. I'm starting a new REST-only service and I'm not sure if I should use Spray again, or start this project on Akka-Http. The product I'm developing will be in production in about 2 months, so timing is a little tricky. Best regards, Jacobus On Wednesday, 11 March 2015 19:38:44 UTC+2, rkuhn wrote: Hi Alexey, we cannot make any hard promises on API compatibility at this point, but we expect things to calm down considerably now. There is one thing that we have not yet implemented which might require the addition of a type parameter to FlowGraph.Builder, but as I said this list is possibly not exhaustive. Regards, Roland 11 mar 2015 kl. 03:55 skrev Alexey Romanchuk alexey.r...@gmail.com javascript:: Patrik, is it any fundamental problems in API right now? Do you plan to break backward compatibility before 1.0? вторник, 10 марта 2015 г., 16:18:35 UTC+6 пользователь Patrik Nordwall написал: We will release 1.0 as soon as possible, but not earlier. We will continue to release frequent milestones until we reach release candidate quality. We don't have an estimate of how long that will take, but I guess it is in the magnitude of a few months. Regards, Patrik On Mon, Mar 9, 2015 at 6:48 PM, Ali Beyad alib...@gmail.com wrote: Dear All, Is there any projected timeframe for releasing version 1.0? After the release of 1.0-M4, I wasn't sure how close Akka was to the official 1.0 release of streams. Thank you, Ali Beyad -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+...@googlegroups.com. To post to this group, send email to akka...@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout. -- Patrik Nordwall Typesafe http://typesafe.com/ - Reactive apps on the JVM Twitter: @patriknw [image: Scala Days] http://event.scaladays.org/scaladays-sanfran-2015 -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+...@googlegroups.com javascript:. To post to this group, send email to akka...@googlegroups.com javascript:. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout. *Dr. Roland Kuhn* *Akka Tech Lead* Typesafe http://typesafe.com/ – Reactive apps on the JVM. twitter: @rolandkuhn http://twitter.com/#!/rolandkuhn -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
[akka-user] Stop shard region gracefully
Hello, how is it possible to stop a shard region and all the shards located on the same node gracefully? Thanks a lot Hannes -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
[akka-user] Stop shard region gracefully
Hello, I am trying to stop my current Cluster Sharding application gracefully. So far I couldn't find a way to stop a shard region and all the shards located on the same node gracefully. What would be a good way to achieve that? Thanks a lot Hannes -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
[akka-user] SslTlsCipherActor: Getting access to MultiStreamInputProcessor$SubstreamOnError
Hi, I'm trying to setup a client using TLS over TCP to connect to a server. The client is using Akka Streams 1.0-M4. At the moment I'm stymied: 1. After a series of debug messages about handshake, wrap and unwrap, the SslTlsCipherActor writes a debug log of ### HS NEED_TASK Running delegated task sun.security.ssl.Handshaker$DelegatedTask@38dcaf04 2. Then, after about 5s I get: Message [akka.stream.impl.MultiStreamInputProcessor$SubstreamOnError] from Actor[akka://chirp-dbus-notifier/deadLetters] to Actor[akka://chirp-dbus-notifier/user/ssl-client-1#1884597281] was not delivered Presumably that SubstreamOnError contains the exception that would give me a clue about what's failing. However, it's not being delivered to my actor and I can't include it in my actor anyway because it's private. So: 1. How do I get access to that exception? 2. Is there a Akka Stream 1.0-M4 example of using TLS over TCP? Many thanks in advance, Edward -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
Re: [akka-user] Stop shard region gracefully
I am interested in that too. So far, I've seen that leaving cluster gracefully with cluster.leave(...), won't cause shard manager to automatically rebalance actors. Instead, it relies on rebalance-interval settings. On Thu, Mar 12, 2015 at 9:41 AM, Hannes Stockner hannes.stock...@gmail.com wrote: Hello, I am trying to stop my current Cluster Sharding application gracefully. So far I couldn't find a way to stop a shard region and all the shards located on the same node gracefully. What would be a good way to achieve that? Thanks a lot Hannes -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout. -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
Re: [akka-user] Stop shard region gracefully
I think you are looking for #16050 https://github.com/akka/akka/issues/16050. I have actually a prototype of that ready so it will be available in Akka 2.4 (or some earlier snapshot/milestone). Cheers, Patrik On Thu, Mar 12, 2015 at 11:28 AM, Dragisa Krsmanovic dragis...@gmail.com wrote: I am interested in that too. So far, I've seen that leaving cluster gracefully with cluster.leave(...), won't cause shard manager to automatically rebalance actors. Instead, it relies on rebalance-interval settings. On Thu, Mar 12, 2015 at 9:41 AM, Hannes Stockner hannes.stock...@gmail.com wrote: Hello, I am trying to stop my current Cluster Sharding application gracefully. So far I couldn't find a way to stop a shard region and all the shards located on the same node gracefully. What would be a good way to achieve that? Thanks a lot Hannes -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout. -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout. -- Patrik Nordwall Typesafe http://typesafe.com/ - Reactive apps on the JVM Twitter: @patriknw [image: Scala Days] http://event.scaladays.org/scaladays-sanfran-2015 -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
Re: [akka-user] Again about tell don't ask and use Actor tell instead future
Hi, *rrodseth*
thank you for link, I read it and try to re-write my code from Futures to
actor 'Tell', but at now example DAO are not async (but I think at now it's
not a problem):
import akka.actor._
import scala.concurrent.duration._
object ExampleActorBoot extends App {
val system = ActorSystem(ExampleActorBoot)
val dao = system.actorOf(Props[DaoActor], dao)
val processor = system.actorOf(Props(new ProcessorActor(test, dao)),
processor)
processor ! ProcessorActor.Process
}
case class TaskId(processorName: String, id: String)
case class TaskDetails(id: TaskId, status: String, json: String)
object DaoActor {
sealed trait DaoActorMessage {
def receiver: ActorRef
}
case class LoadTaskIds(processorName: String, receiver: ActorRef) extends
DaoActorMessage
case class LoadTaskDetails(id: TaskId, receiver: ActorRef) extends
DaoActorMessage
case class DeleteTask(id: TaskId, receiver: ActorRef) extends DaoActorMessage
case class MarkSuccess(id: TaskId, receiver: ActorRef) extends DaoActorMessage
}
class DaoActor extends Actor with ActorLogging {
import com.episodetracker.boot.DaoActor._
override def receive = {
case LoadTaskIds(processorName, receiver) =
val result = 1 to 4 map {
case i =
TaskId(processorName, i.toString)
}
receiver ! result
case LoadTaskDetails(id, receiver) =
val result = TaskDetails(id, pending, JSON)
receiver ! result
case DeleteTask(id, receiver) =
receiver ! true
case MarkSuccess(id, receiver) =
receiver ! true
}
}
object ProcessorActor {
sealed trait ProcessorActorMessage
case object Process extends ProcessorActorMessage
private case class TaskComplete(id: TaskId, isSuccess: Boolean) extends
ProcessorActorMessage
private case class ProcessingComplete(result: Seq[(TaskId, Boolean)]) extends
ProcessorActorMessage
}
class ProcessorActor(name: String, dao: ActorRef) extends Actor with
ActorLogging {
import com.episodetracker.boot.ProcessorActor._
override def receive = {
case Process =
dao ! DaoActor.LoadTaskIds(name, context.actorOf(Props(new
TasksProcessorResultActor(
case ProcessingComplete(result) =
sender() ! PoisonPill
log.info(sProcessing complete with results [$result])
case unknown =
log.error(sUnknown message [$unknown])
}
val emptyResult = Seq[(TaskId, Boolean)]()
class TasksProcessorResultActor extends Actor {
context.setReceiveTimeout(3 seconds)
override def receive = default(0, emptyResult)
def default(awaitResults: Int, completedTasks: Seq[(TaskId, Boolean)]):
Receive = {
case tasks: Seq[TaskId] =
tasks.foreach {
case task =
context.actorOf(Props(new TaskProcessorActor(task)))
}
context become default(tasks.size, emptyResult)
case TaskComplete(id, isSuccess) =
val results = completedTasks :+ ((id, isSuccess))
if (results.size == awaitResults) {
context.parent ! ProcessingComplete(results)
} else {
context become default(awaitResults, results)
}
case unknown =
log.error(sUnknown message [$unknown])
}
}
class TaskProcessorActor(taskId: TaskId) extends Actor with ActorLogging {
override def receive = default
dao ! DaoActor.LoadTaskDetails(taskId, self)
def default: Receive = {
case taskDetails: TaskDetails =
taskDetails.status match {
case success =
dao ! DaoActor.DeleteTask(taskDetails.id, self)
context.become(waitDaoResponse)
case _ =
//TODO: do some processing
dao ! DaoActor.MarkSuccess(taskDetails.id, self)
context.become(waitDaoResponse)
}
case unknown =
log.error(sUnknown message [$unknown])
}
def waitDaoResponse: Receive = {
case true =
context.parent ! TaskComplete(taskId, true)
case false =
context.parent ! TaskComplete(taskId, false)
}
}
}
I think it's not perfect and if code in //TODO: do some processing need
some DAO call, actor code with become go to hell.
What in second sample can be corrected or improved?
With best regards, Vladimir.
среда, 11 марта 2015 г., 17:14:21 UTC+3 пользователь rrodseth написал:
Hi Vladimir
I have had good luck with using per-request actors as in the Net-a-porter
activator template, and a replyTo:ActorRef in the messages sent to my
DbActor (which still talks to a DAO)
Some more detail in this thread
http://qnalist.com/questions/5581605/using-tell-or-ask-with-akka-persistence-and-rest
On Tue, Mar 10, 2015 at 2:55 PM, Владимир Морозов green...@gmail.com
javascript: wrote:
Hi All,
this question about design, for example now I have some process that load
task list and try process each task, it looks like this:
import scala.concurrent.{ExecutionContext, Future}
case class TaskId(processorName:
Re: [akka-user] Re: Problems using a Remote Actor
A quick update, a notable difference in my sources is that I call the actorSelection(...) but from my client actor system, and not from another actor (in the usual way contect.actorSelection(...) find in many documents): ActorSelection selection = systemClient.actorSelection(remoteBasePath + remoteActorName); could be this ? If yes, are there some good reason to not do the call in this way ? Should I open an issue in Akka (to fix it, or at least to clarify this in docs) ? Sorry for all these questions, but I'm really trying to better understand (and simplify if possible) this ... Thanks, Sandro Il giorno giovedì 12 marzo 2015 11:07:00 UTC+1, Sandro Martini ha scritto: Hi Jim, thanks for the help ... Have you followed all of the instructions for interacting with remote actors here: http://doc.akka.io/docs/akka/2.3.9/scala/remoting.html ? I follow this because I'm using the Java version of the API: http://doc.akka.io/docs/akka/2.2.5/java/remoting.html and them the updated (more/better info) http://doc.akka.io/docs/akka/2.3.9/java/remoting.html Note that I have the same behavior from 3 different pc (Windows and Linux) and with different patch levels on JDK 7 (but all recent), so it's something related to code ... Just for info, my project is made with Gradle 2.3, so to try it its' enough to checkout with git, and run it with Gradle (if already installed, or its wrapper will download it), and all dependencies are downloaded at first run, I made it to make it simple for anyone to run/try it. From the error it appears to me that: 1. You are attempting to create a remote actor from a peer yes, but I didn't read for problems with it in documentation ... 2. This peer isn't able to properly interact with [akka.tcp:// RemoteActorSystem-Client@127.0.0.1:55003] yes but from the error/warning message it's not clear what is the real problem ... Are you using a Remote or Clustered ActorRefProviders on both nodes? no, I'm using normal akka.remote.RemoteActorRefProvider Is your remote transport properly configured to run on those ports? Is it starting properly? Is your networking configured to allow access to those ports (software firewalls on a node have burned me in the past). yes, I see that the related port is open, with an OS diagnostic tool Does the RemoteActorSystem-Client AS exist on that port, or is some other AS running there? nothing special is running There's a lot to look at, but it ought to work once everything is wired properly. yes, but all seems ok, so really I think it's a problem in my code Maybe look for a working example (from Activator) and then adapt the working thing to your needs? maybe this is a good option, but my sample here is even to try to use Akka from a (little) different point of view (and tools like Gradle) ... maybe even Akka documentation could be improved, at least to better explain some pitfalls on exposing actors with remoting ... just as a suggestion for Akka team. One of goals for my test project is even to share/improve some knowledge on using Akka from other tools (mainly the Groovy/Gradle/Grails stack) with others :-) . Thanks a lot for the help. Sandro -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
Re: [akka-user] Cluster: Multiple leaders / ReachableMember not propagated / ...?
On Tuesday, March 10, 2015 at 4:17:10 PM UTC+1, Patrik Nordwall wrote: What was puzzling me was: Cluster Node [akka.tcp://...Node4...] - Marking node(s) as REACHABLE [Member(address = akka.tcp://Node1..., status = Up)] but looking at the code revealed that this is only an notification that Node4 thinks that Node1 is reachable again. The ReachableMember event is fired when all thinks it is reachable again, and that will not happen until Node3 is back in business or removed. So...while being still in this status - Node 1: MemberStatus=Up, Leader = 1 / Unreachable = Node 3 - Node 2: MemberStatus=Up, Leader = 2 / Unreachable = Node 1, 3 - Node 4: MemberStatus=Up, Leader = 2 / Unreachable = Node 1, 3 ...is it a problem for Node 2 and Node 3 if Node 1 has the cluster singleton? I mean, from the state above, i suppose Node 1 will be able handle application tasks (which require cluster singleton actions) perfectly. But what about Node 2 and Node 4? Will they use the cluster singleton of Node 1, even if they have Node 1 in their Unreachable List? Thanks, Michael -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
Re: [akka-user] Cluster: Multiple leaders / ReachableMember not propagated / ...?
On Tue, Mar 10, 2015 at 4:38 PM, michaels michael.schr...@atos.net wrote: On Tuesday, March 10, 2015 at 4:17:10 PM UTC+1, Patrik Nordwall wrote: What was puzzling me was: Cluster Node [akka.tcp://...Node4...] - Marking node(s) as REACHABLE [Member(address = akka.tcp://Node1..., status = Up)] but looking at the code revealed that this is only an notification that Node4 thinks that Node1 is reachable again. The ReachableMember event is fired when all thinks it is reachable again, and that will not happen until Node3 is back in business or removed. So...while being still in this status - Node 1: MemberStatus=Up, Leader = 1 / Unreachable = Node 3 - Node 2: MemberStatus=Up, Leader = 2 / Unreachable = Node 1, 3 - Node 4: MemberStatus=Up, Leader = 2 / Unreachable = Node 1, 3 ...is it a problem for Node 2 and Node 3 if Node 1 has the cluster singleton? I mean, from the state above, i suppose Node 1 will be able handle application tasks (which require cluster singleton actions) perfectly. But what about Node 2 and Node 4? Will they use the cluster singleton of Node 1, even if they have Node 1 in their Unreachable List? Yes, cluster singleton decisions are not based on unreachability information at all. /Patrik Thanks, Michael -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout. -- Patrik Nordwall Typesafe http://typesafe.com/ - Reactive apps on the JVM Twitter: @patriknw [image: Scala Days] http://event.scaladays.org/scaladays-sanfran-2015 -- Read the docs: http://akka.io/docs/ Check the FAQ: http://doc.akka.io/docs/akka/current/additional/faq.html Search the archives: https://groups.google.com/group/akka-user --- You received this message because you are subscribed to the Google Groups Akka User List group. To unsubscribe from this group and stop receiving emails from it, send an email to akka-user+unsubscr...@googlegroups.com. To post to this group, send email to akka-user@googlegroups.com. Visit this group at http://groups.google.com/group/akka-user. For more options, visit https://groups.google.com/d/optout.
